1 | initial version |

Based on personal experience, there are a couple things you can try:

Set

**maximum velocity impulse (max_vel)**to 0 to enforce pure position correction with no added momentum from interpenetration correction. [Source]

Alternatively, set to max_vel to some small, non-zero value (e.g 1-10 m/s to limit the added momementum from position correctionSet a

**finite minimum allowable interpenetration depth (min_depth)**of 0.001 [m] (more generally a small value relative to the scale of the colliding objects. [Source]

If that doesn't make a difference, you may also try **decreasing the physics simulation step size** to get a better solution as explained here.

2 | No.2 Revision |

Based on personal experience, there are a couple things you can try:

Set

**maximum velocity impulse (max_vel)**to 0 to enforce pure position correction with no added momentum from interpenetration correction. [Source]

Alternatively, set to max_vel to some small, non-zero value (e.g 1-10 m/s to limit the added momementum from position correctionSet a

**finite minimum allowable interpenetration depth (min_depth)**of 0.001 [m] (more generally a small value relative to the scale of the colliding objects. [Source]

If that doesn't make a difference, you may also try **decreasing the physics simulation step size** to get a better solution as explained here~~.~~. This will descrease numerical instability but also slow down your simulation.

3 | No.3 Revision |

Based on personal experience, there are a couple things you can try:

Set

**maximum velocity impulse (max_vel)**to 0 to enforce pure position correction with no added momentum from interpenetration correction. [Source]

Alternatively, set to max_vel to some small, non-zero value (e.g 1-10 m/s to limit the added momementum from position correctionSet a

**finite minimum allowable interpenetration depth (min_depth)**of 0.001 [m] (more generally a small value relative to the scale of the colliding objects. [Source]

If that doesn't make a difference, you may also try **decreasing the physics simulation step size** to get a better solution as explained here. This will descrease numerical instability but also slow down your simulation.

4 | No.4 Revision |

Based on personal experience, there are a couple things you can try:

**maximum velocity impulse (max_vel)**to 0 to enforce pure position correction with no added momentum from interpenetration correction. [Source]

Alternatively, set to max_vel to some small, non-zero value (e.g 1-10 m/s to limit the added momementum from position correction**finite minimum allowable interpenetration depth (min_depth)**of 0.001 [m] (more generally a small value relative to the scale of the colliding objects. [Source]

If that doesn't make a difference, you may also try **decreasing the physics simulation step size** to get a better solution as explained here. This will ~~descrease ~~reduce numerical instability but also slow down your simulation.

5 | No.5 Revision |

Based on personal experience, there are a couple things you can try:

Set

**maximum velocity impulse (max_vel)**to 0 to enforce pure position correction with no added momentum from interpenetration correction. [Source]

Alternatively, set to max_vel to some small, non-zero value*(e.g 1-10*to limit the added momementum from position correction~~m/s~~m/s)**finite minimum allowable interpenetration depth (min_depth)**of 0.001 [m] (more generally a small value relative to the scale of the colliding objects. [Source]

If that doesn't make a difference, you may also try **decreasing the physics simulation step size** to get a better solution as explained here. This will reduce numerical instability but also slow down your simulation.

Copyright Askbot, 2010-2011. Content on this site is licensed under a Creative Commons Attribution Share Alike 3.0 license.